Banca de QUALIFICAÇÃO: HILTON THALLYSON VIEIRA MACHADO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : HILTON THALLYSON VIEIRA MACHADO
DATE: 16/10/2026
TIME: 09:00
LOCAL: Remoto
TITLE:

App2Car: A Mobile Edge AI Platform for Vehicular Data Acquisition and Analysis


KEY WORDS:

Connected Vehicles; Edge AI; Mobile Application; Small Language Models (SLMs); On-Device Inference.


PAGES: 45
BIG AREA: Engenharias
AREA: Engenharia Elétrica
SUMMARY:

The continuous evolution of embedded electronic systems and in-vehicle communication networks has contributed to the increasing volume and diversity of telemetry data provided by Electronic Control Units (ECUs). This work proposes developing a mobile application that communicates with the vehicle to collect vehicular sensor data and provide real-time analytical insights to users, integrating embedded artificial intelligence models that predict fuel type, classify roads, detect anomalies, and analyze driver behavior. Furthermore, the application features an offline chatbot powered by local Small Language Models (SLMs) combined with Retrieval-Augmented Generation (RAG) to enable querying of vehicle owners’ manuals. The solution was developed using the Flutter framework, interfacing with ELM327 adapters via the On-Board Diagnostics (OBD-II) protocol over Bluetooth Low Energy to acquire vehicle data. Case studies were conducted to evaluate the impact of variables such as OBD adapter types, smartphone models, and vehicle electronic architecture on data acquisition latency and model execution times. These results provide experimental evidence regarding the feasibility of integrating embedded artificial intelligence techniques, OBD-II data acquisition, and local language models on mobile devices for real-time vehicular applications.


COMMITTEE MEMBERS:
Interno - 2885532 - IVANOVITCH MEDEIROS DANTAS DA SILVA
Externo ao Programa - 2249146 - CARLOS MANUEL DIAS VIEGAS - UFRNExterno ao Programa - 3374361 - JEAN MARIO MOREIRA DE LIMA - UFRNExterna à Instituição - MARIANNE BATISTA DINIZ DA SILVA - UFAL
Notícia cadastrada em: 28/08/2026 06:06
SIGAA | Superintendência de Tecnologia da Informação - (84) 3342 2210 | Copyright © 2006-2026 - UFRN - sigaa05-producao.info.ufrn.br.sigaa05-producao